The Development and Statistical Evaluation of a Recruit Training Performance Test.
Abstract
The Recruit Performance Test (RPT) was developed to meet a widely recognized need for an accurate measurement of achievement in learning the military/psychomotor skills in recruit training. It was hoped that the test would be of special value in assessing the achievement of Category IV personnel, who are thought to be penalized by the present emphasis upon verbal/academic testing. The RPT, composed of items in the recruit training curriculum which were the most amenable to performance testing, was administered to recruits in their fourth through tenth weeks of training. Despite intensive efforts to develop a wide variety of appropriate test items, the following deficiencies were found: Most recruits were able to perform at a very high level, thus rendering the test too easy for effective discrimination; Although inter-rater reliabilities for the test were acceptable, K-R20 reliabilities were marginal and alternate form reliabilities were low. These deficiencies would preclude the use of the RPT for comparisons among individual personnel; however, the test may prove to be useful as an index of performance differences between groups. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1973
- Accession Number
- AD0760307
Entities
People
- Bernard Rimland
- Charles H. Cory
- Edmund D. Thomas
- James Hysham
Organizations
- United States Naval Research Laboratory